OBS. 33.—­According to this gentleman’s notion of things, there is, within the little circle of the word what, a very curious play of antecedent parts and parts relative—­a dodging contra-dance of which that and that which, with things which, and so forth.  Thus:  “When what is a compound relative you must always parse it as two words; that is, you must parse the antecedent part as a noun, and give it case; the relative part you may analyze like any other relative, giving it a case likewise.  Example:  ‘I will try what (that which) can be found in female delicacy.’  Here that, the antecedent part of what, is in the obj. case, governed by the verb ‘will try;’ which, the relative part, is in the nom. case to ‘can be found.’  ’I have heard what (i.e. that which, or the thing which) has been alleged.’ “—­Kirkham’s Gram., p. 111.  Here, we sec, the author’s “which-that” becomes that which, or something else.  But this is not a full view of his method.  The following vile rigmarole is a further sample of that “New Systematick Order of Parsing,” by virtue of which he so very complacently and successfully sets himself above all other grammarians:  “‘From what is recorded, he appears, &c.’ What is a comp. rel. pron. including both the antecedent and the relative, and is equivalent to that which, or the thing which.—­Thing, the antecedent part of what, is a noun, the name of a thing—­com. the name of a species—­neuter gender, it has no sex—­third person, spoken of—­sing. number, it implies but one—­and in the obj. case, it is the object of the relation expressed by the prep. ‘from,’ and gov. by it:  RULE 31. (Repeat the Rule, and every other Rule to which I refer.) Which, the relative part of what, is a pronoun, a word used instead of a noun—­relative, it relates to ‘thing’ for its antecedent—­neut. gender, third person, sing, number, because the antecedent is with which it agrees, according to RULE 14. Rel. pron. &c. Which is in the nom. case to the verb ’is recorded,’ agreeably to RULE 15. The relative is the nominative case to the verb, when no nominative comes between it and the verb.”—­Kirkham’s Gram., p. 113.

OBS. 34.—­The distinction which has been made by Murray and others, between etymological parsing and syntactical—­or, between that exercise which simply classifies and describes the words of a sentence, and that which adds to this the principles of their construction—­is rejected by Kirkham, and also by Ingersoll, Fuller, Smith, Sanborn, Mack, and some others, it being altogether irreconcilable with their several modes of confounding the two main parts of grammar.  If such a distinction is serviceable, the want of it is one of the inherent faults of the schemes which they have adopted.  But, since “grammar